Re: [open-cobol-list] feature request: EVALUATE: partial-expression

[Thomas Biehler]

On Wednesday 18 February 2004 19:55, William M. Klein wrote:
> I am NOT at all positive about this, but I *think* that the MF extension is
> slightly different from the ISO 2002 Support.  As I recall - MF allows:
>

If you are negative, perhaps because you can't recall,
that is NOT the topic here. ;-)

I don't know, i can only guess.
Do you have no access to MF-COBOL Language Reference,
or do you have no time for reading?
(Perhaps, because you are too busy 
 of writing usefull hints to others !?)

So, for your convenience, here is the relevant clipping
from the MF-COBOL Language Reference that i
have explicitly referenced in my autotest-example before!

MF-COBOL Language Reference: EVALUATE-STATEMENT: Syntax Rules (7)
==========================================================
  If partial-expression-1 or partial-expression-2 is specified as a 
  selection object, it must correspond to a selection subject
  that is an identifier, a literal or an arithmetic expression.
  Partial-expression-1 and partial-expression-2 must be a 
  sequence of COBOL words which, when following the 
  corresponding selection subject, make the resultant text a
  valid conditional expression. 
==========================================================

NOTE:
 ...  to a selection subject, that is an identifer, a literal or 
      an arithmetic expression ... 

>   Evaluate ABC >
--------------->  HERE you fail !
---------------> "ABC >"  is  also NOT  allowed in MF-COBOL!
--------------->  Operator ">"  => "ABC > X"  is a relation condition!
>      When "X"
>             ....
    
>
> In other words, I think that MF allows the expression to be "broken"
> anywhere - while ISO 2002 REQUIRES that only the "subject" be in the
> EVALUATE statement - and the rest be in the WHEN.
>

MF does NOT allows broken expressions (at this point ;-)  ! 

> Bottom-Line:
>   Before implementing this, one needs to "check" out the exact supported
> syntax in each definition.

Do you know anyone ?   :-)

Always thankful for good contributions.

Thomas Biehler